[Webkit-unassigned] [Bug 40376] [Qt] Sometimes I see pixel dust when the transform is changing rapidly with Qt graphics layers

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Jun 16 03:40:47 PDT 2010


https://bugs.webkit.org/show_bug.cgi?id=40376





--- Comment #8 from Jędrzej Nowacki <jedrzej.nowacki at nokia.com>  2010-06-16 03:40:47 PST ---
(In reply to comment #6)
> (In reply to comment #3)
> > (From update of attachment 58555 [details] [details])
> > Actually I should close this bug report as invalid, as it is not possible to reproduce the problem, please provide us a test case :-). Good luck!
> >
> > > +        No new tests. (OOPS!)
> > It is possible to add a new autotest?
> I don't have an autotest, it was clearly a corner case that had the pixels left uncovered. I very much doubt I'll get a case to cover it - but can try later. Can you point me to a autotest that does pixel comparisons?
I can't as you didn't provide a test case, I can't reproduce it. I'm not sure what you want to fix, but maybe you should look at graphicsview autotest (part of Qt).

> > Spaces at the end of the line
> Why do you want spaces at the end of the line?
Sorry, my message wasn't clear. I don't want spaces at the end of line, but you shouldn't change it, as it is not related to the pixel dust problem. Actually the rule is more general a patch should touch only relevant code and only fix one issue.

> > > +#if 0
> > If you believe that this code in unneeded you should remove it.
> This was for Noam to experiment with as I think my fix removes the need for the above "hack" as they are solving a similar problem.

> > > +#if 1
> Agree the #if 1 here is unnecessary - its a piece of code I never triggered so I left in Noam's "hack" to work around the issue since I couldn't confirm whether it was still necessary. My hope was that Noam would test if it was still necessary in the test that required it and possibly remove it.
Great, so we need Noam's expertise :-)
Preferably, we avoid #if 0/1. If you feel that a code could be removed then you can remove it, if code have to stay then you shouldn't touch it. If you need to share some experimental code, try to do it in other branch unless it is something really big. WebKit code base is huge and it is difficult to maintain it so lets try to keep it simple :-)

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.


More information about the webkit-unassigned mailing list